(Log in to disable ads.)

So many speaker choices. In terms of frequency response curves, comparing some Celestion curves with Jensen curves indicates that various guitar drivers have pretty much the same pattern. Low on bass, dip at 400 Hz, high at 2KHz and up, roll off at 4 – 5 KHz. On the other hand, Hi-Fi has other requirements, with a “flat” response through most of the audio range as in the first figure. The effect of the venue itself is illustrated in the second graph, where room resonances boost some frequencies, and damp others.

Fig.4 KEF Blade Two, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with complex sum of nearfield responses plotted below 300Hz.
Read more at https://www.stereophile.com/content/kef-blade-two-loudspeaker-measurements#cisfBifajp1YXI4S.99

Fig.7 KEF Blade Two, spatially averaged, 1/6-octave response in JA's listening room (red); and of DALI Rubicon 8 (blue).
Read more at https://www.stereophile.com/content/kef-blade-two-loudspeaker-measurements#cisfBifajp1YXI4S.99

I would assume PA speakers are related to Hi-Fi in order to faithfully reproduce whatever is amplified.
In comparison, the 10” Greenback response curve peaks around 2 – 4 KHz, and rolls off quite a bit before you hit the low E on the guitar. If we take -3 dB to be half the noise, seven dB roll-off is significant.

Pondering this, I felt the irrational urge to compare the supplied frequency response curves with real-world guitar frequencies (uhm, well, reproduced through a guitar speaker unfortunately). This requires studio-quality recording and Hi-Fi quality playback equipment. Using Audacity, with the laptop microphone, I recorded some guitar. Both acoustic and electric, tuned to A=440Hz. I used two electric guitars, both at the same settings through a 40 Watt 12 Inch valve combo, recording single coils neck and bridge, humbuckers neck and bridge. Play one string, all the others undamped, play a note, damp the other strings. And so forth. Audacity frequency plots were eye-opening. There are many harmonics, some stronger than the fundamental. My recording setup is not good enough to do this rigorously, but it shows trends.
The steel string acoustic showed some weird resutls: Plying the Bass E, the peak frequency in the spectrum plot is 163 Hz, not the 82 Hz expected. Harmonics run up to 4755 Hz. To test, I generated a tone at the required fundamental frequency, and played guitar track and tone together. Yup, my guitar recorded that frequency loud and clear. What should I make of Audacity’s analyses? A strong peak at 239 Hz (B?) and several other peaks indicate the other strings starting to enjoy this party. Luthiers, comment please.
Playing the treble E shows no surprises, a strong peak at the fundamental frequency (330 Hz), with harmonics at 997 Hz and 1651 Hz peaking about 5 dB louder (?), and the rest of the plot filled with harmonics in multiples of the fundamental frequency, although they tend to go sharp.

Fretting the treble E at the 12th fret, Uhm yes. The fundamental at 664 Hz dominates. With a strong peak 2/3 rd’s of this frequency down (A flat). At 2651 Hz (E) and 3325 Hz (A flat) peaks just 5 dB and 7 dB softer than the fundamental frequency catches the eye. Harmonics continue up past 6 kHz. Here too, the other strings want to join the fun.

The 12th fretted treble E string may just indicate the origin of that “icepick”: The harmonics run into a region that is very sharp, right in that speaker frequency response peak. No, damping the rest of the strings when fretting this note did not take away the off-note peaks, the other notes still shine through.

This really requires studio equipment. Any volunteers?

Hi Modulator

Thank you for being the guy bringing actual data to the table --- the world needs more people like you. I agree with everything you wrote above.

So the real question in your plot above is what the frequency response of your laptop mic is. If you can get your hands on a microphone with a published frequency response (basically anything from Shure: http://blog.shure.com/how-to-read-a-microphone-frequency-response-chart/), you will be able to correct your chart above to represent what your ears are actually hearing.

Basically you will need to generate an EQ that is opposite to the freq response of the microphone, i.e. if the mic is cutting 100Hz by 3dB, you have to boost 100Hz by 3dB.

I am just guessing here, but I suspect the reason you are seeing your first harmonic on the low E (163Hz) being louder than the fundamental (82Hz) is because of the poor low end response of your laptop mic. Once you correct for that, the fundamental should stick out its head again ?

I agree, studio equipment will make for a better comparison, especially if you can use things with published frequency response. Have fun!

  • V8 replied to this.

    klaasvakie So the real question in your plot above is what the frequency response of your laptop mic is.

    ? Thought the same - leads me in to a noobie question. Let's say we've got a SM57, how would you mic up a cab to get the 'cleanest' (I'm not sure what word to use here) response for the purposes of Modulator's experiment?

      V8 Thought the same - leads me in to a noobie question. Let's say we've got a SM57, how would you mic up a cab to get the 'cleanest' (I'm not sure what word to use here) response for the purposes of Modulator's experiment?

      The way to correct for the microphones frequency response is to apply the inverse EQ, see pic

      Black is the frequency response of the Shure SM57, red is the EQ you need to apply to get a flat frequency response into the recording (excuse my paint skillz)

      There is one thing I am unsure of, according to Shure there is a proximity effect when the mic is close to the source, check halfway down this page (http://blog.shure.com/how-to-read-a-microphone-frequency-response-chart/) where they talk about the Beta 58A. It looks like the proximity effect causes a prominent bass boost, but I'm not sure how much of a role that plays when mic'ng up a cab.

      I guess if it were me (and in the absence of fancy test equipment) I would keep alternating between playing some notes on the the instrument and listening to the recording while playing with the EQ until they sound identical. You have to trust your ears at some point ?

      • V8 likes this.

        @klaasvakie - thanks, all very interesting!

        I'll use the summation on ledgernote as my takeaway : "The goal is always to "get it right at the source." And that means don't take close miking too literally. Maintain at least 12 inches of distance between your source and your microphone unless you specifically intend to draw upon the bass response of tight proximity"

          I am not an acoustic engineer/expert, and finding one with time to do this sort of thing for free to post here is not going to happen. Fortunately the professionals do this and info is available. What I would have liked to do (time permitting) is to do this in an acoustically "neutral" room, with good recording equipment, decent software, and Hi-Fi playback. To test the instruments will be an interesting project in itself. "Testing" speaker choice will have me recording the guitar (acoustic with microphone, electric will have to go straight from the pickup to the recording medium), do the equalization filters to "simulate" various guitar speakers (in "pure" form, no effects from room, cabinet, amplifier), and play back through the Hi-Fi to see if one can actually hear a difference in response. In no way does this relate to the real-world sound you will get from a cabinet. But it might help to decide whether you fork out money for the green backed thing or a cheap speaker, or to decide between two or three options. All the time remembering that you are doing this to find a setup which pleases your ears.

          I'm sure many thesis'es have been done on this, and surely the manufacturers do this all the time. As I said, I needed to know. And also see that some more questions pop up. As for now, I will have to use my equipment, and have frustrating fun.

            I need a better microphone, without spending a nice playable SX's worth of money. I was thinking about an USB mike for the laptop. Any advice?

            modulator

            Where do you plan to play? How much damage does it need to withstand? If you're going to take it live, you'll want something sturdy. I have a Shure SM58 that I use for live gigs and it is built like a beast. I also have an AKG C214 that's great for vocals.

            Slightly off topic, but how do you evaluate the frequency response of a pair of speakers? Is there any software tool that will do it?

                13 days later

                Norman86 posted in another thread (uhm, moderators, how does one reference here?):

                2 x 8" Celestion drivers in a custom cab, 200W power handling...
                Celestion TF0818 drivers..

                oh, and it's WAY too loud. I love it! ?
                And yes, it's mine! ?

                Two Celestion TF 0818 drivers for a 200 Watt 2 x 10” cabinet. These are Hi-Fi drivers, low-midrange, specified frequency response of 70 – 6000 Hz, sensitivity 94 dB. Rated power 100 Wrms. The interweb is full of horror stories about using Hi-Fi speakers for guitar, most of the arguments against it not well thought through, defeating themselves. The consensus seems to be, plug your guitar amp into a Hi-Fi cabinet (driver) and you will cause Earth’s atmopsphere to ignite in a supernova fireball, destroying life as we know it. Some even venture to say that guitar speakers are full on all the time, whereas audiophile type (read PA?) speakers have a on-off (loud-soft) job, so run cooler, etc. Sure. Listening to some heavy music, that Hi-Fi driver has to handle not only one guitar, but drums, bass, keyboards, more guitars, screaming (sorry, vocals) etc. And live gigs through a PA is not for sissy speakers, but they are full range flat response (like Hi-Fi) cabinets and run hard all the time. While the guitar amp/speaker is run just enough to monitor, being miked to the deck etc.

                Note the TF0818’s frequency response curve, off-axis: It drops off nicely at 3KHz. The on-axis peak at 3 – 4 KHz is much like a guitar driver, so I would venture a guess that adding a “beam blocker” over the dome will result in a pleasing guitar speaker without shrill painfull treble. And the speaker will have a more “controlled” sound as well, if the user does not try to drive it into flubbering destruction.

                Indeed, Celestion now markets a speaker for modelling amps, a true full range flat response speaker, the F12-X200. 60 Hz – 20 KHz. Look at that response curve:

                It has a 97 dB sensitivity rating too. I would merely remove the on-axis compression driver to get the ideal “guitar” speaker. Note the catalogue blurp implies that this speaker allows you full control over your sound. So where are the Hi-Fi type speaker detractors now? The times (and requirements) have changed.

                • V8 replied to this.

                  I feel I have to put my 5c in here so here we go, note these are generalisations, differt speakers may have different "ideas" I am not looking at HF units for this discussion.

                  Hi Fi speakers
                  Generally need to be fairly compact, have extended bass and flat frequency response, to achieve this sensitivity at anywhere from 85 to just over 90 dB is generally way too low for guitar or PA applications. Power handling is usually smaller as it is just not needed in the average living room.
                  Car speakers
                  Will be of lower impedance, usually 4 Ohm, to match the requirement of the car amp which runs into a lower impedance to provide more power with less current draw and therefore less drain on the battery. They need to survive the harsher enviroment in a car, perform in less than ideal placement positions and get loud enough to overcome the background noise level.
                  PA or pro audio speakers
                  The bigger more powerful units will have very stiff cones and heavy voicecoils. Bass response will be limited in order to get decent efficiency and most will need a properly designed ported or otherwise loaded cabinet. They respond slowly and with less attack compared to a guitar speaker. Power handling will be higher when driven by high continuous power amps but this will reduce substatially on a bass or guitar amp. Less costly units will have more in common with guitar speakers such as lighter cones and suspensions and smaller voice coils but still need good continuous power handling. This improves sensitivity at the expense of bass response but who needs 40Hz from their guitar?
                  Bass speakers
                  Obviously need to get lower down than guitar speakers but still need attack and good sensitivity. Power handling needs to be high but additionally they ney need to cope with sudden large inputs and cone movements from percussive playing styles.
                  Guitar speakers
                  This is where the "dark arts" of speaker design come in. guitar speakers are designed to produce a specific tonal quality and to break up in a particular manner when overdriven. Frequency response graphs are virtually meaningless in illustrating the results you will get. Guitar speakers sensitivity is often 97dB or more, response drops off sharply below 80Hz and they respond quickly to small nuances in the signal. Smaller cone drivers such as 6 or 8 inch will have much in common with the less costly pa drivers discussed above but 12" speakers are another story.

                  • V8 likes this.

                  The new modeling type amps provide a different requirement with a need to tonally influence the signal as little as possible. For the audience these are usually fed to the FOH PA system and for the guitarist back to the monitors.
                  In the "old days" most of the guitar sound reaching the audience came from the on stage guitar cabs, think double stack (or more) 4 x 12 cabs and 100W valve heads. Oh, and incidently, guitarists with damaged hearing. This is no longer the case and the crowd is used to hearing guitar sound through the PA. Not so the guitarist who now has to listen to wedge monitors which do not have the same characteristics as a guitar cab.

                  This is why Celestion have developed the new Copperback and the F12-X200. The intention is to keep the "feel" of a guitar speaker while providing a fairly neutral platform for the modelling gear to do it's work. I feel the F12 will be more suited to studio work and the Copperback to live but time will tell. There would not be much point in removing the HF driver, just use a Copperback instead.

                  • V8 likes this.

                  Another reason why you would want a full range speaker in a modelling amp is that the cab simulation is usually part of the modelling, so playing it through a guitar speaker will effectively apply the speaker frequency response twice.

                  I have a Line6 Amplifi practice amp which explicitly has a full range speaker (but no published frequency response). This allows me to load different cab-sims in software and still reproduce them as they should be. This also means that the Line-in/Bluetooth inputs get reproduced accurately and only the guitar bits get their frequencies chomped.

                  Obviously there can be some lively debate about the ability of the Line6 to model a cab accurately in software, but the principle should stand.

                  • V8 likes this.

                  modulator Norman86 posted in another thread (uhm, moderators, how does one reference here?):

                  Just cut n paste the url from the thread and it'll pop up as a hyperlink your post. Also you can 'tag' someone by putting an @ in front of their username (E.g. @modulator means you'll get a notification from both this reply and the tag in this sentence).

                  The info that @JGP and @klaasvakie are sharing is gold - thanks guys!

                      Good information on speaker design here. OK, more questions pop up as to Hi-Fi, where sickening amounts of cash gets spent buying equipment which allows the fortunate owner to be "at the concert" 24/7/365. From that I assumed that the speakers have the ability to reproduce the tiniest nuance and sound of good (this is important) recordings. The lightest pick attack, the smallest squeek, the subtle hammer-on, etc. Which apparently happens.

                      Another thing is that the only reference I (and most other people alive) have of the classic rock sound, the true blues tone, what Clapton sounded like in studio when he recorded whatever, Gilmour's magic tone, is derived from Hi-Fi. We have recordings to listen to, and form opinions of what we like, but that is possibly not what it sounded like when played.

                      I never had good guitars, good amps, good speakers. I have never heard/played any . I cannot blindfolded say "Fender/Marshall/Vox" and ID speaker cabinets, strings, guitars. My only reference is my Hi-Fi. The few live shows I attended was coloured by the PA 'Mix". So what does a guitar really sound like? Likely why I am not getting the "tone".

                        So to add to this... Patrick Quilter of Quilter amps and QSC audio has many, many speakers at his disposal.
                        I'm pretty sure he knew what he was doing pairing his amps with the "hifi" speakers.

                        As I said in the other thread, if you're in JHB, you're more than welcome to come check my cab out and listen to how it responds.

                          "OK, more questions pop up as to Hi-Fi, where sickening amounts of cash gets spent buying equipment which allows the fortunate owner to be "at the concert" 24/7/365. From that I assumed that the speakers have the ability to reproduce the tiniest nuance and sound of good (this is important) recordings."

                          Yes exactly. I spent most of my younger years in the hi fi business when it was still fresh and good fun. Good Hi Fi speakers, like any other speaker, are a compromise of factors but are trying to achieve a diffent goal to PA or others.
                          They must:
                          Be reasonably sized
                          get loud enough to fill the listning room
                          have a broad, flat, smooth frequency response
                          look good
                          and be priced reasonably.
                          There are of course exceptions, I have seen very large hi fi speakers, heard very loud units and also some awful units with spitty top end and boomy or "one note" bass.

                          The objective of a good hi fi system should be to reproduce the source as closely as possible AND allow the listner to enjoy the result. Some do a very good job but something is always missing. How many times have you heard music from another room or outside a venue and you just know it's live before you walk in? Hi Fi just doesn't do that for me.

                          "So what does a guitar really sound like? Likely why I am not getting the "tone"."

                          In my opinion a guitar should sound like, well, the way you want it to.

                            Write a Reply...